Marshall Schor (JIRA | 23 Oct 19:39 2014
Picon

[jira] [Resolved] (UIMA-4060) improve int set utility impls


     [
https://issues.apache.org/jira/browse/UIMA-4060?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Marshall Schor resolved UIMA-4060.
----------------------------------
    Resolution: Fixed

> improve int set utility impls
> -----------------------------
>
>                 Key: UIMA-4060
>                 URL: https://issues.apache.org/jira/browse/UIMA-4060
>             Project: UIMA
>          Issue Type: Bug
>          Components: Core Java Framework
>    Affects Versions: 2.6.1SDK
>            Reporter: Marshall Schor
>            Assignee: Marshall Schor
>            Priority: Minor
>             Fix For: 2.6.1SDK
>
>
> The PositiveIntSet impl has a bug in the ordered set iterator (not incrementing an index in a loop).  In
addition, Improve this also by eliminating expensive operations on bit Sets (such as computing the
number of elements in the set), and augmenting the IntBitSet to allow a potentially much smaller
footprint, by supporting an "offset" for the use case where there's a lot of fairly "dense" members,
clustered at an offset from 0.  Increase the hysteresis for switching among alternative impls in
PositiveIntSet. Add (approximate) tracking for IntHashSet of mostNegative as well as most positive. Do
some other minor cleanup - making things appropriately private / final.  Add / update test cases.
(Continue reading)

Marshall Schor (JIRA | 23 Oct 19:19 2014
Picon

[jira] [Created] (UIMA-4060) improve int set utility impls

Marshall Schor created UIMA-4060:
------------------------------------

             Summary: improve int set utility impls
                 Key: UIMA-4060
                 URL: https://issues.apache.org/jira/browse/UIMA-4060
             Project: UIMA
          Issue Type: Bug
          Components: Core Java Framework
    Affects Versions: 2.6.1SDK
            Reporter: Marshall Schor
            Assignee: Marshall Schor
            Priority: Minor
             Fix For: 2.6.1SDK

The PositiveIntSet impl has a bug in the ordered set iterator (not incrementing an index in a loop).  In
addition, Improve this also by eliminating expensive operations on bit Sets (such as computing the
number of elements in the set), and augmenting the IntBitSet to allow a potentially much smaller
footprint, by supporting an "offset" for the use case where there's a lot of fairly "dense" members,
clustered at an offset from 0.  Increase the hysteresis for switching among alternative impls in
PositiveIntSet. Add (approximate) tracking for IntHashSet of mostNegative as well as most positive. Do
some other minor cleanup - making things appropriately private / final.  Add / update test cases.

--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Marshall Schor (JIRA | 21 Oct 23:20 2014
Picon

[jira] [Commented] (UIMA-4059) add-to-indexes and checking for incorrect key modifications


    [
https://issues.apache.org/jira/browse/UIMA-4059?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14179125#comment-14179125
] 

Marshall Schor commented on UIMA-4059:
--------------------------------------

The tracking of which FSs are added/removed to/from indices needs to be "per view", since it's perfectly OK
to add the same FS to multiple views (if it is not a subtype of AnnotatorBase).  The testing of feature
modifications has to check if the FS is indexed in any view - a potentially more expensive operation.  The
overheads involved in add/remove to/from index are such that the additional check probably is too small
to measure, but the overhead for testing if a FS is in any index for every setting of Feature Values that are
in keys might be excessive.  Since the most common pattern is to set several features for a new FS at once, it
may pay to have a one-element cache of the last FS that was found to not be in any index, or to have this check be
separately omittable (like Java asserts -
  you could turn it on if you have some issue like unexpected index behavior.)

> add-to-indexes and checking for incorrect key modifications
> -----------------------------------------------------------
>
>                 Key: UIMA-4059
>                 URL: https://issues.apache.org/jira/browse/UIMA-4059
>             Project: UIMA
>          Issue Type: Improvement
>          Components: Core Java Framework
>    Affects Versions: 2.6.0SDK
>            Reporter: Marshall Schor
>            Assignee: Marshall Schor
>            Priority: Minor
(Continue reading)

Marshall Schor (JIRA | 21 Oct 22:27 2014
Picon

[jira] [Created] (UIMA-4059) add-to-indexes and checking for incorrect key modifications

Marshall Schor created UIMA-4059:
------------------------------------

             Summary: add-to-indexes and checking for incorrect key modifications
                 Key: UIMA-4059
                 URL: https://issues.apache.org/jira/browse/UIMA-4059
             Project: UIMA
          Issue Type: Improvement
          Components: Core Java Framework
    Affects Versions: 2.6.0SDK
            Reporter: Marshall Schor
            Assignee: Marshall Schor
            Priority: Minor
             Fix For: 2.6.1SDK

Address the issue raised in Jiras UIMA-3399 and UIMA-4049, as follows: Add (under control of a Java jvm
defined param - for backwards compatibility in unexpected edge cases, or if it is determined this has too
big a performance/space cost) a set which tracks if a FS has been added/removed-to/from-the-indexes
that has fast O(1) access. Change behavior of add-to-indexes when a FS has already been added to not add it
again. Change behavior of feature structure value setting to throw an exception if the feature being set
is used as a key in any index, and the FS has been added-to-the-indexes.

This should address issues raised in both JIRAs.

--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Marshall Schor (JIRA | 21 Oct 21:17 2014
Picon

[jira] [Resolved] (UIMA-4056) fix JSON tests which do XML compares to use XMLUnit


     [
https://issues.apache.org/jira/browse/UIMA-4056?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Marshall Schor resolved UIMA-4056.
----------------------------------
    Resolution: Fixed

> fix JSON tests which do XML compares to use XMLUnit
> ---------------------------------------------------
>
>                 Key: UIMA-4056
>                 URL: https://issues.apache.org/jira/browse/UIMA-4056
>             Project: UIMA
>          Issue Type: Improvement
>          Components: Core Java Framework
>    Affects Versions: 2.6.1SDK
>            Reporter: Marshall Schor
>            Assignee: Marshall Schor
>            Priority: Minor
>             Fix For: 2.6.1SDK
>
>
> The JSON tests have a mode where they generate xmi/xml serializations as well, just for extra coverage. 
THe compare of these against expected can't be a string "equals" because different XML parsers,
transformers, etc. do ordering of name-spaces and formatting of empty elements differently.  Use
instead the XMLUnit package (available from Maven Central - license OK) (thanks to suggestion from
[~rec]), to check for "equality" instead that allows for these differences.

--
(Continue reading)

Jaroslaw Cwiklik | 17 Oct 15:43 2014
Picon

[VOTE] Release UIMA-DUCC 1.1.0 RC5

Hi,

the Apache UIMA DUCC v1.1.0 RC5 is ready for voting.

DUCC stands for Distributed UIMA Cluster Computing. DUCC is a cluster
management system providing tooling, management, and scheduling facilities
to automate the scale-out of applications written to the UIMA framework.
Core UIMA provides a generalized framework for applications that process
unstructured information such as human language, but does not provide a
scale-out mechanism. UIMA-AS provides a scale-out mechanism to distribute
UIMA pipelines over a cluster of computing resources, but does not provide
job or cluster management of the resources. DUCC defines a formal job model
that closely maps to a standard UIMA pipeline. Around this job model DUCC
provides cluster management services to automate the scale-out of UIMA
pipelines over computing clusters.

There were two issues fixed since RC4

- Manually generated JIRA report to include in the release. Maven changes
plugin only pulls in the first 100 jiras. Ducc 1.1.0  has 189 bugs so about
half were missing.

 - Minor documentation fixes in ducc book.

https://issues.apache.org/jira/issues/?jql=project%20%3D%20UIMA%20AND%20fixVersion%20%3D%20%221.1.0-Ducc%22%20ORDER%20BY%20key%20ASC

The Maven artifacts are here:
https://repository.apache.org/content/repositories/orgapacheuima-1040/org/apache/uima/

The source and binary zip/tars are here:
(Continue reading)

Jerry Cwiklik (JIRA | 16 Oct 19:20 2014
Picon

[jira] [Closed] (UIMA-4057) Ducc's JIRA report should be manually generated due to JIRA bug


     [
https://issues.apache.org/jira/browse/UIMA-4057?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jerry Cwiklik closed UIMA-4057.
-------------------------------
    Resolution: Fixed

Added issuesFixed with generated content into svn
Disabled maven changes plugin in top level pom
Updated assembly script to copy issuesFixed folder to bin and source distro

> Ducc's JIRA report should be manually generated due to JIRA bug
> ---------------------------------------------------------------
>
>                 Key: UIMA-4057
>                 URL: https://issues.apache.org/jira/browse/UIMA-4057
>             Project: UIMA
>          Issue Type: Bug
>          Components: DUCC
>    Affects Versions: 1.0.0-Ducc
>            Reporter: Jerry Cwiklik
>            Assignee: Jerry Cwiklik
>             Fix For: 1.1.0-Ducc
>
>
> It looks like Apache JIRA has a bug returning a max of 100 jiras when creating a jira report. Ducc 1.1.0 has a
total of 188 jiras, so 88 are not in the report. Even when running a query directly in JIRA, one gets only the
first 100.
> A workaround is to run the query manually in JIRA, let it return the first 100 than change query parameter
(Continue reading)

Burn Lewis (JIRA | 16 Oct 17:43 2014
Picon

[jira] [Closed] (UIMA-4058) Fix docs for job descriptor overrides


     [
https://issues.apache.org/jira/browse/UIMA-4058?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Burn Lewis closed UIMA-4058.
----------------------------

> Fix docs for job descriptor overrides
> -------------------------------------
>
>                 Key: UIMA-4058
>                 URL: https://issues.apache.org/jira/browse/UIMA-4058
>             Project: UIMA
>          Issue Type: Bug
>          Components: DUCC
>    Affects Versions: 1.1.0-Ducc
>            Reporter: Burn Lewis
>            Assignee: Burn Lewis
>            Priority: Minor
>             Fix For: 1.1.0-Ducc
>
>
> Overrides are blank not comma separated

--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Burn Lewis (JIRA | 16 Oct 17:42 2014
Picon

[jira] [Resolved] (UIMA-4058) Fix docs for job descriptor overrides


     [
https://issues.apache.org/jira/browse/UIMA-4058?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Burn Lewis resolved UIMA-4058.
------------------------------
       Resolution: Fixed
    Fix Version/s: 1.1.0-Ducc

Fixed docs description of descriptor overrides & a few typos

> Fix docs for job descriptor overrides
> -------------------------------------
>
>                 Key: UIMA-4058
>                 URL: https://issues.apache.org/jira/browse/UIMA-4058
>             Project: UIMA
>          Issue Type: Bug
>          Components: DUCC
>    Affects Versions: 1.1.0-Ducc
>            Reporter: Burn Lewis
>            Assignee: Burn Lewis
>            Priority: Minor
>             Fix For: 1.1.0-Ducc
>
>
> Overrides are blank not comma separated

--
This message was sent by Atlassian JIRA
(Continue reading)

Burn Lewis (JIRA | 16 Oct 17:14 2014
Picon

[jira] [Created] (UIMA-4058) Fix docs for job descriptor overrides

Burn Lewis created UIMA-4058:
--------------------------------

             Summary: Fix docs for job descriptor overrides
                 Key: UIMA-4058
                 URL: https://issues.apache.org/jira/browse/UIMA-4058
             Project: UIMA
          Issue Type: Bug
          Components: DUCC
    Affects Versions: 1.1.0-Ducc
            Reporter: Burn Lewis
            Assignee: Burn Lewis
            Priority: Minor

Overrides are blank not comma separated

--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Jerry Cwiklik (JIRA | 16 Oct 15:30 2014
Picon

[jira] [Created] (UIMA-4057) Ducc's JIRA report should be manually generated due to JIRA bug

Jerry Cwiklik created UIMA-4057:
-----------------------------------

             Summary: Ducc's JIRA report should be manually generated due to JIRA bug
                 Key: UIMA-4057
                 URL: https://issues.apache.org/jira/browse/UIMA-4057
             Project: UIMA
          Issue Type: Bug
          Components: DUCC
    Affects Versions: 1.0.0-Ducc
            Reporter: Jerry Cwiklik
            Assignee: Jerry Cwiklik
             Fix For: 1.1.0-Ducc

It looks like Apache JIRA has a bug returning a max of 100 jiras when creating a jira report. Ducc 1.1.0 has a
total of 188 jiras, so 88 are not in the report. Even when running a query directly in JIRA, one gets only the
first 100.

A workaround is to run the query manually in JIRA, let it return the first 100 than change query parameter
value of tempMax to 200. 

Temporarily disable changes plugin in ducc's top pom to stop generating partial report. Add manually
generated report to svn so that the build can pick this up and include it in the tar and zip files.

--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


Gmane